00:44
akheir has quit [Quit: Leaving]
00:46
shahrzad has quit [Quit: Leaving]
00:57
bita_ has quit [Ping timeout: 260 seconds]
01:56
hkaiser has quit [Quit: bye]
02:23
kale[m] has quit [Ping timeout: 265 seconds]
02:34
kale[m] has joined #ste||ar
06:23
kale[m] has quit [Ping timeout: 240 seconds]
06:23
kale[m] has joined #ste||ar
06:43
jaafar has quit [Quit: Konversation terminated!]
10:19
kale[m] has quit [Read error: Connection reset by peer]
10:20
kale[m] has joined #ste||ar
10:21
<
gonidelis[m] >
I cannot understand what is this build fail about
13:46
hkaiser has joined #ste||ar
14:00
akheir has joined #ste||ar
14:10
jaafar has joined #ste||ar
14:22
RostamLog has joined #ste||ar
14:23
kale[m] has joined #ste||ar
15:01
bita_ has joined #ste||ar
15:29
<
gonidelis[m] >
repost: I cannot understand what this build fail is about
15:33
K-ballo has quit [Ping timeout: 240 seconds]
15:35
K-ballo has joined #ste||ar
15:38
<
hkaiser >
gonidelis[m]: it's one of the know problems we have, unrelated to your work
15:44
<
gonidelis[m] >
hkaiser: thanks
15:56
<
gonidelis[m] >
hkaiser: so should I assume that everything is fine with the pr by now?
15:56
<
gonidelis[m] >
good
15:56
<
hkaiser >
if that's the only failure
16:50
<
gonidelis[m] >
hkaiser: I have removed the projection documentation comments from hpx::transform doxygen part
16:50
<
gonidelis[m] >
I only keep them for hpx::ranges::transform
16:51
<
gonidelis[m] >
is that correct?
17:11
<
gonidelis[m] >
hkaiser: yt?
17:11
<
gonidelis[m] >
this overload does not have `InIter1`
17:12
<
gonidelis[m] >
I think it means `begin(rng)`
17:12
<
gonidelis[m] >
should I leave it as is ?
17:14
<
gonidelis[m] >
ahhh.... actually I don't think that such an overload even exists in std
17:14
<
gonidelis[m] >
hkaiser: ^^
17:27
<
hkaiser >
it's range_iterator<Rng>::type
17:27
<
hkaiser >
good catch
17:53
ramses[m] has joined #ste||ar
18:22
kale[m] has quit [Ping timeout: 240 seconds]
18:22
kale[m] has joined #ste||ar
18:54
<
gonidelis[m] >
any help on how I should treat that?
19:01
<
gonidelis[m] >
hkaiser:
19:06
kale[m] has quit [Ping timeout: 260 seconds]
19:07
kale[m] has joined #ste||ar
19:12
<
hkaiser >
gonidelis[m]: isn't that the same as for the other algorithms?
19:26
<
gonidelis[m] >
hkaiser: ok you are right
19:26
<
gonidelis[m] >
my bad
19:27
<
gonidelis[m] >
What do we mean by "signature" ?
19:30
<
K-ballo >
usually the type of a function, or a function type representing the types in a call expression
19:30
<
K-ballo >
what's the context?
19:31
<
K-ballo >
that's some unusual wording
19:31
<
gonidelis[m] >
what's the suggestion
19:32
<
K-ballo >
the function type, or the function call operator of the given callable, or..., does not need to have const
19:32
<
K-ballo >
`[](const Type &a) -> Ret {}` will do, but so will `[](Type &a) -> Ret {}`
19:32
<
K-ballo >
uhm, that sounds wrong
19:33
<
K-ballo >
I'm guessing it's supposed to say "does not need to have `const Type &`"
19:33
<
K-ballo >
ah, no, it's saying `[](Type a) -> Ret {}` will work too
19:34
<
K-ballo >
the whole phrasing is... off
19:34
<
K-ballo >
it also says predicate, but it seems this is not actually a predicate?
19:35
<
K-ballo >
it is definitely not a predicate
19:36
<
K-ballo >
hah, that odd wording is copied verbatim from cppreference
19:37
<
K-ballo >
not really.. it's the same, isn't it? it has all the same mistakes
19:38
<
K-ballo >
you want to say that `f`, which is not a predicate (may be an operation if you wish to give it some sort of name), is callable with a `const Type &`
19:38
<
K-ballo >
btw, verbatim from cppref explains the odd const placement
19:40
<
K-ballo >
except `Type` is a magic placeholder here..
19:40
<
K-ballo >
it's callable with whatever the iterator yields when dereferenced
19:41
kale[m] has quit [Ping timeout: 240 seconds]
19:41
<
K-ballo >
don't get stuck on it, looks like a pervasive defect
19:42
kale[m] has joined #ste||ar
19:43
<
gonidelis[m] >
K-ballo: cool thanks
19:47
<
gonidelis[m] >
K-ballo: why f is not a predicate?
19:48
<
K-ballo >
predicates are boolean
19:48
<
K-ballo >
they serve as conditions
19:49
<
K-ballo >
copy_if gets a predicate, transform gets an ordinary operation
19:55
<
K-ballo >
in some form or other, yeah
20:03
<
gonidelis[m] >
the other two overloads accordingly:
20:04
<
gonidelis[m] >
the other one* overload actually
20:04
<
gonidelis[m] >
...two in total
20:04
<
gonidelis[m] >
K-ballo: what do you think?
20:05
<
gonidelis[m] >
(feel free to comment as a review if you want to)
20:05
<
gonidelis[m] >
hkaiser: we wait for the tests and we are set ;)
20:06
<
gonidelis[m] >
next stop: `min_max`
20:08
<
K-ballo >
can't give any useful feedback when the entire wording is broken
20:08
<
K-ballo >
I could tell you the text is incorrect, but we knew that already
20:08
<
K-ballo >
let it be for now, it's going to require an entire overhaul for all algorithms' docs
20:10
<
gonidelis[m] >
K-ballo: the thing that `f` does is that it takes the range and applies its functionality on it, right?
20:11
<
K-ballo >
rephrase?
20:11
<
K-ballo >
the algorithm operates on the range, f only operates on an element/pair of elements
20:11
<
gonidelis[m] >
it applies some transformation on the given range
20:11
<
gonidelis[m] >
ok yeah
20:12
<
gonidelis[m] >
does `f` define the kind of transformation that is going to take place?
21:11
kale[m] has quit [Ping timeout: 265 seconds]
21:11
kale[m] has joined #ste||ar
21:18
kale[m] has quit [Ping timeout: 265 seconds]
21:31
kale[m] has joined #ste||ar
21:38
kale[m] has quit [Ping timeout: 265 seconds]
21:55
kale[m] has joined #ste||ar
22:41
kale[m] has quit [Ping timeout: 264 seconds]
22:41
kale[m] has joined #ste||ar
23:10
kale[m] has quit [Quit: Quit]